Question

我有一个程序可以扫描目录中的文本文件，遍历每一行并根据每行中的前缀进行解析。该程序充当提取器，从前缀“Hxx”上的Base64字符串中提取tif图像。当程序从“Hxx”行获取图像时，它只是删除原始文件。

我想要做的是保持行“TXA”的过滤条件，但不是将“Hxx”行上的字符串转换为图像并删除文件，我想保留文件的全部内容。基本上，只使用程序根据行“TXA”的条件来解析和过滤文本文件。

我知道在我的case循环的foreach“TXA”中，我需要将整个文件保存到内存流中，以便在程序结束时重新写入文件。我现在还不确定。

非常感谢任何帮助。

        /// <summary>
    /// This method will open, read and parse out the image file and save it to disk.
    /// </summary>
    /// <param name="fileName"></param>
    /// <returns></returns>
    static bool ParseHFPFile(string inputFileName, string outputFileName)
    {
        List<MemoryStream> tiffStreamList = new List<MemoryStream>();

        // 1. grab file contents.
        string fileContents = ProgramHelper.GetFileContents(inputFileName);
        if (string.IsNullOrEmpty(fileContents))
        {
            return false; // errors already raised.
        }
        Log("[O] ", false);

        // 2. split file contents into a string array.
        string[] fileContentStringList = fileContents.Split('\r');
        if (fileContentStringList.Length == 0)
        {
            Log(" ERROR: Unable to split file contents on [CR] character.");
            return false;
        }

        // 3. loop through the file lines and parse each "section" based on it's prefix.
        string mrn = string.Empty;
        string dos = string.Empty;
        string imgType = string.Empty;
        foreach (string line in fileContentStringList)
        {
            if (string.IsNullOrEmpty(line))
            {
                continue;
            }
            string prefix = line.Substring(0, 3);
            switch (prefix)
            {
                case "MSH":
                    break;
                case "EVN":
                    break;
                case "PID":
                    mrn = line.Split('|')[3].Split('^')[0];

                    break;
                case "PV1":
                    dos = line.Split('|')[44];
                    if (!String.IsNullOrWhiteSpace(dos))
                    {
                        dos = dos.Substring(0, 8);
                    }
                    break;
                case "TXA":
                    imgType = line.Split('|')[2].Split('^')[0];
                    if (imgType == "EDH02" || imgType == "EDORDH")
                    {
                        Log("[NP]");

                        return true;
                    }
                    break;
                case "OBX":
                    break;
                case "Hxx":
                    // 0 - Hxx
                    // 1 - page number
                    // 2 - image type
                    // 3 - base64 encoded image.

                    // 1. split the line sections apart based on the pipe character ("|").
                    string[] hxxSections = line.Split('|');
                    byte[] decodedImageBytes = Convert.FromBase64String(hxxSections[3].Replace(@"\.br\", ""));

                    // 2. create a memory stream to store the byte array.
                    var ms = new MemoryStream();
                    ms.Write(decodedImageBytes, 0, decodedImageBytes.Length);

                    // 3. add the memory stream to a memory stream array for later use in saving.
                    tiffStreamList.Add(ms);
                    break;
                case "Z":
                    break;
            }
        }
        Log("[P] ", false);

        // 4. write the memory streams to a new file.
        ImageCodecInfo icInfo = ImageCodecInfo.GetImageEncoders().Single(c => c.MimeType == "image/tiff");
        System.Drawing.Imaging.Encoder enc = System.Drawing.Imaging.Encoder.SaveFlag;
        var ep = new EncoderParameters(1);

        // 5. create references to the EncoderValues we will use
        var ep1 = new EncoderParameter(enc, (long)EncoderValue.MultiFrame);
        var ep2 = new EncoderParameter(enc, (long)EncoderValue.FrameDimensionPage);
        var ep3 = new EncoderParameter(enc, (long)EncoderValue.Flush);

        string newOutputFilePath = Path.GetDirectoryName(outputFileName) + @"\";
        string newOutputFileName = newOutputFilePath + mrn + "_" + dos + ".dat";
        bool success = false;
        int suffix = 1;
        while (!success)
        {
            if (File.Exists(newOutputFileName))
            {
                newOutputFileName = newOutputFilePath + mrn + "_" + dos + "_" + suffix + ".dat";
            }
            else
            {
                success = true;
            }
            suffix++;
        }
        Log(string.Format("[NewFile: {0}] ", Path.GetFileName(newOutputFileName)), false);

        var strm = new FileStream(newOutputFileName, FileMode.OpenOrCreate, FileAccess.ReadWrite);
        System.Drawing.Image pages = null;

        int frame = 0;
        int pageCount = tiffStreamList.Count;

        Log("[WT:", false);
        try
        {
            foreach (MemoryStream m in tiffStreamList)
            {
                if (frame == 0)
                {
                    ep.Param[0] = ep1;
                    pages = Image.FromStream(m, false, false);
                    pages.Save(strm, icInfo, ep);
                }
                else
                {
                    ep.Param[0] = ep2;
                    if (pages != null)
                        pages.SaveAdd(Image.FromStream(m, false, false), ep);
                }

                if (frame == pageCount - 1)
                {
                    ep.Param[0] = ep3;
                    if (pages != null)
                        pages.SaveAdd(ep);
                }

                frame++;
                Log(".", false);
                //m.Close();
                m.Dispose();
            }
            Log("]");

            return true;
        }
        catch (Exception ex)
        {
            Log(" EXCEPTION: " + ex.Message + Environment.NewLine + ex.StackTrace);
            return false;
        }
        finally
        {
        }
    }

Answer 1

我很高兴您成功完成了任务。如果我的评论帮助您解决了问题，我会将其作为答案发布，以便您可以接受它。

因此，您可以直接创建两个List<string>，而不是寻找更复杂的方法：

List<string> TXA = new List<string>();
List<string> FilesForDelete = new List<string>();

然后在您的代码中，取决于您要对文件执行的操作：

if (fileIsTXA)
{
  TXA.Add(fileName);
}
else
{
  FilesForDelete.Add(fileName);
}

稍后您可以使用这两个列表来提取文件名，并随意使用它们。

需要扫描目录中的文本文件并根据循环结果进行过滤

1 个答案: